---
title: README
emoji: 📊
colorFrom: blue
colorTo: indigo
sdk: static
pinned: false
---

# InfoVis: Unified Slide Image Generation and Editing

One 9B model, fine-tuned from FLUX.2-klein-base-9B, for text-to-slide generation, template-conditioned generation, multi-image grounded generation, and instruction-based slide editing.

<p align="center">
  <img src="https://huggingface.co/InfoVis/InfoVis/resolve/main/assets/overview.jpg" width="90%">
</p>

| | |
|---|---|
| 📄 Paper | [arXiv:2610.16784](https://arxiv.org/abs/2610.16784) |
| 🌐 Project page | [vectorspacelab.github.io/InfoVis](https://vectorspacelab.github.io/InfoVis/) |
| 💻 Code | [VectorSpaceLab/InfoVis](https://github.com/VectorSpaceLab/InfoVis) |
| 🤗 Model | [InfoVis/InfoVis](https://huggingface.co/InfoVis/InfoVis) |
| 🚀 Demo | [Try InfoVis online](https://huggingface.co/spaces/sabulin/InfoVis): generate and edit slides in the browser |
| 📊 Benchmark | [InfoVis/VisSlideBench](https://huggingface.co/datasets/InfoVis/VisSlideBench): 1,527 manually screened cases in four tracks |
| 📦 Training data | [InfoVis/Data](https://huggingface.co/datasets/InfoVis/Data): ≈15M records for continual pretraining and SFT |